Bug 337391

Summary:	mount -a gives errors if gfs is already mounted
Product:	Red Hat Enterprise Linux 5	Reporter:	Lutz Lange <llange>
Component:	gfs-utils	Assignee:	Robert Peterson <rpeterso>
Status:	CLOSED CURRENTRELEASE	QA Contact:	GFS Bugs <gfs-bugs>
Severity:	low	Docs Contact:
Priority:	low
Version:	5.0	CC:	kzak
Target Milestone:	---	Keywords:	Reopened
Target Release:	---
Hardware:	i386
OS:	Linux
Whiteboard:
Fixed In Version:	RHEL5.1	Doc Type:	Bug Fix
Doc Text:		Story Points:	---
Clone Of:		Environment:
Last Closed:	2007-11-28 16:30:21 UTC	Type:	---
Regression:	---	Mount Type:	---
Documentation:	---	CRM:
Verified Versions:		Category:	---
oVirt Team:	---	RHEL 7.3 requirements from Atomic Host:
Cloudforms Team:	---	Target Upstream Version:
Embargoed:

Description Lutz Lange 2007-10-18 08:32:15 UTC

Description of problem:
gfs filesystem is already mounted.
mount -a is executed and writes :


Version-Release number of selected component (if applicable):
Enterprise Linux 5.0 
gfs-utils-0.1.11-3.el5

How reproducible:
mount your gfs filesystem and execute mount -a
  
Actual results:
error message

Expected results:
no output, no action

Comment 3 Robert Peterson 2007-10-22 13:28:37 UTC

The bug was closed by accident so I'm reopening.

Comment 6 Robert Peterson 2007-10-22 22:02:20 UTC

Ah, herein lies the problem:

"all paths in /etc/mtab have to be canonicalized absolute paths"

Because we're using lvm2 and device mapper, the canonicalized absolute
path to /dev/TAFT_CLUSTER/TAFT_CLUSTER0 is actually
/dev/mapper/TAFT_CLUSTER-TAFT_CLUSTER0.  I lost access to this cluster,
but I recreated it on another.  After I modified /etc/mtab to have the
full canonicalized absolute path, mount -a found it as well.

In version 5.1, the mount helper puts the resolved path into mtab,
so we don't have the problem there.  I'll backtrack it from here.

Comment 7 Robert Peterson 2007-10-23 18:48:52 UTC

I haven't backtracked this yet, but the fact remains: the problem does
not occur in rhel5.1.  Seeing as how it's too late to get anything into
5.0.z anyway, rather than waste any more of my time trying to dig through
code, I'd like to just close this bug as CURRENTRELEASE.  Is that alright?

Comment 8 Lutz Lange 2007-11-28 15:44:26 UTC

alright with me

Comment 9 Robert Peterson 2007-11-28 16:30:21 UTC

Closing CURRENTRELEASE as discussed above.